Copy




September 2015

2

Confidence in the Evidence from Reviews of Qualitative Research 



Welcome to our newsletter!  In this edition you will find out about recent work on dissemination bias in qualitative research, the CERQual events to look out for at the Cochrane Colloquium and information on a recently published review that applied CERQual

CERQual and Cochrane
 
Are you planning to attend the Cochrane Colloquium in Vienna between Saturday 3rd and Wednesday 7th October, 2015? If so, we hope to see you there! We particularly hope to see you at one or more of our CERQual featured sessions. These are listed in the righthand column. You can find out more about these sessions by clicking the 'read more' link.  

The Cochrane Collaboration, and specifically its annual meeting at the Cochrane Colloquium, has been a critical component of the ongoing evolution of the CERQual approach. The CERQual team uses interaction with the wider systematic review and decision-making community at the Colloquium as an opportunity to develop, test and refine the CERQual principles and procedures. Widespread consultation and friendly, constructive criticism is as central to the development and uptake of an approach to grading the certainty of review findings as it is to adoption of quality assessment checklists or reporting standards.
The CERQual team is presenting the latest version of the approach at an oral presentation between 14:00-15:30 on Sunday 4th October entitled Assessing how much confidence to place in findings from qualitative evidence syntheses: a new version of the GRADE-CERQual tool (O5.2 – Room Lehar 2). Then two days later the latest version of CERQual is showcased in more detail at a similarly-titled workshop, 11:00-12:30 on Tuesday 6th October: Assessing how much confidence to place in findings from qualitative evidence syntheses: a new version of the GRADE-CERQual approach (W63 – Room Lehar 3). With the presentations scheduled for proximate rooms on the Ground Floor, albeit two days apart, you hardly need to move – unless, of course, you find yourself tempted by Vienna’s famous patisseries or the strains of the Blue Danube. So get yourself "slow slow quick quick slow" to our CERQual feature events and share a part in shaping the ongoing direction of this exciting initiative.   
 
Dr Andrew Booth, Co-Convenor of the Cochrane Qualitative and Implementation Methods Group 
 

Using the Latest Version of CERQual in a Recent Systematic Review 
 
Our review team used the latest version of the CERQual approach in a recently published systematic review entitled “The mistreatment of women during childbirth in health facilities globally: a mixed-methods systematic review”. This review was commissioned by the World Health Organization’s Department of Reproductive Health and Research as part of the quality of care research portfolio.​

Two members of the review team developed the summary of qualitative findings table and conducted the CERQual assessments. The reviewers used a structured template in Excel to track the decision process and developed an evidence profile, which included all information on the judgements underlying the CERQual assessment. Using the CERQual approach challenged the review team to ensure that the structure of each review finding was a reasonable representation of the phenomenon of interest and was an accurate reflection of the data from the contributing studies. Developing the summary of qualitative findings tables also helped to present the findings from the qualitative synthesis in a clear and user-friendly format. The summary of qualitative findings table is currently being used as the backbone to develop a labour observation and community survey tool to measure how women are treated during childbirth. Access the article HERE

Meghan Bohren, Department of Reproductive Health and Research, WHO Geneva. Department of Population, Family and Reproductive Health, Johns Hopkins Bloomberg School of Public Health 

Our team from the University of Sheffield, supported by the wider CERQual-DEQS Subgroup, is in the final stages of a mapping review to establish what has been published on dissemination effects in qualitative research. Our findings so far indicate a relative paucity of empirical evidence focusing on publication bias in qualitative research. The data also shows that many challenges faced by qualitative manuscripts are introduced during editorial review.​

“Dissemination effects”, as in the Subgroup’s full name - Dissemination Effects in Qualitative Synthesis - refers to a series of biased decisions potentially made during the design, implementation and publication of research. Such biases, when present, restrict or prevent the dissemination of research thereby resulting in wastage. In the specific context of the CERQual approach these biases pose a threat to the coherence, adequacy and relevance of review findings as well as undermining the methodological quality of the review and its constituent studies.
Publication bias, within the wider remit of ‘dissemination effects’, has been well researched with regards to quantitative research, but relatively little attention is paid to the challenges faced specifically by qualitative research. Following a systematic approach, we conducted searches on multiple databases, selected the most relevant studies and coded the extracted data against a conceptual framework depicting the stages of the dissemination pathway.
Authors from several disciplines report common obstacles such as the absence of definitive evaluation criteria for qualitative research and unaccommodating or unsympathetic editorial policies. Furthermore, non-prejudicial extrinsic factors are also reported to influence the dissemination of qualitative research.
Further detail of findings from this mapping review will be contained in a forthcoming publication – when and if published!

Dr Sarange Nyakang’o and Dr Andrew Booth (A.Booth@sheffield.ac.uk) , University of Sheffield
Dissemination Bias in Qualitative Research from the Perspective of Researchers, Reviewers and Editors 
 
An international project group, led by researchers from Cochrane Germany, is exploring  the phenomenon of dissemination bias in qualitative research. Over 1000 respondents completed an online survey which explored the extent of non-publication of qualitative research studies, and awareness and understanding of the issue of dissemination bias in qualitative research. Preliminary results will be presented at the 23rd Cochrane Colloquium (Poster Session 2 on Wednesday, 07th October) in Vienna.​

Background
Dissemination bias in qualitative research is not well researched but likely highly prevalent and particularly relevant to consider in systematic reviews incorporating qualitative evidence and evidence syntheses of qualitative studies. The trustworthiness of synthesized qualitative findings and conclusions is dependent on appropriate consideration of possible dissemination bias in analysis and interpretation of findings. Insights into dissemination bias in qualitative research can also inform the further development and refinement of the CERQual approach.
Methods
The target audience for the survey included qualitative researchers, peer-reviewers and editors of scientific journals. Their experiences that may have contributed to reducing or facilitating dissemination bias in qualitative research was of interest. The survey also wanted to find out about opinions towards measures which could be employed to decrease possible dissemination bias.
The survey was online for 24 days in August/September 2015 and invitations as well as reminders were sent via mailing lists, professional networks and to researchers who published a report of a qualitative study recently and whose e-mail addresses were identified through a search in OVID Medline.

Many thanks to all readers who participated in our survey and shared their experiences, attitudes and perceptions. We appreciate your effort very much.
Contact:
Ingrid Toews: toews@cochrane.de
Joerg Meerpohl: Meerpohl@cochrane.de

Ingrid Toews, Cochrane Germany, Medical Center - University of Freiburg, Germany 
 
 
 



•
•
•